1. Field of the Invention
The application generally relates to a method for allocating network addresses for network subscribers in a segmented network having a plurality of subnetworks. In addition, the application relates to a router component for performing the method. In particular the application relates to the allocation of network addresses in segmented networks that have different subnetworks that are connected to a busbar system via routing.
2. Related Art
Networks having such a network topology are frequently used for machine installations in order to shield machine-internal subnetworks having sensors and actuators connected thereto from the busbar system to a large extent and thereby to allow machine operation that is both as safe as possible and interference-free. The busbar system is used to network the machines to one another and to other devices, such as an engineering system and/or user interfaces (HMI=Human Machine Interface) for the observation and operation of the machines. In segmented networks of this kind, there frequently is a requirement for direct access to automation-engineering devices in the separate subnetworks, for example, for diagnosis purposes.
Often, operational requirements mean that the entire network is embodied as a separate network. As a result, there is an absence of a default router on a topmost level. In alternative cases, in which there is a default router present, uninterrupted further operation of the network may be required even if the default router fails. In both cases, a stable internal address position is desired even without prior planning and/or manual configuration.
Since the overall network is not flat, but rather is separated into subnetworks, even a busbar system requires each network subscriber connected thereto to have knowledge about all the routes into the subnetworks. This affects programmable logic controllers for machines, an engineering system and/or user interfaces (and also other devices that are directly networked thereto).
It is also desirable for the network topology described to be able to be put into operation without difficulty, and also to be easily extended or changed later as well, for example when a further machine is installed, preferably without particular specialist IT (IT=Information Technology) and network knowledge. Finally, it is desirable to be able to use configuration-free network operation of this kind within parts of larger networks in the future, for example, within production cells as parts of larger production installations.
Known approaches to implementing network separation while being able to reach all network subscribers simultaneously, are via application routing and IP routing (IP=Internet Protocol).
In the case of application routing, the subnetworks, which may be embodied using different kinds of technologies, are used to build a shared (new) network layer with its own network addressing. In this case, such application networks are planned in advance using an engineering tool, for example, so that the available routes can be computed therefrom and loaded into the individual applications. In this case, the concept of application routing is static for some applications. In-situ adjustments for the installation of machines or the extension of existing machinery are not envisaged and always require the overall project, in which the changes need to be planned in advance and from which the amended route data then needs to be loaded into the network subscribers. However, the overall project is generally not available in situ because of practical considerations, and the personnel often does not have the engineering tool and the necessary specialist knowledge for operating it.
From IP routing, users expect more flexibility and continuous and open, standardized communication. However, the architecture of the IP is largely governed by the needs of general IT and by Internet Service Providers (ISP). However, these instances of use do not feature the above-described network topology of segmented separate networks in practice. Accordingly, there is a lack of automatism particularly in the still widely used IPv4 (=Internet Protocol version 4), in order to be able to put these special network topologies into operation, and later extend them, preferably without configuration and without specialist knowledge.